Factors affecting work ability index among polish nurses working in hospitals – A prospective observational survey

Abstract Aims To assess the ability to work of Polish nurses by age groups. Background The ability to work is widely discussed in the literature in the context of nurses' productivity; thus, it is necessary to identify the ability to work when facing an increasing demand for services. Methods The observational study involved 349 professionally active nurses aged 46.9 ± 9.7 years, with a length of service of 23.5 ± 9.6 years. The Work Ability Index (WAI) was used to assess the nurses' ability to work. Results The ability to work decreases with age (r s = −0.324, p < .000) and with seniority (r s = −0.257; p < .000). Nurses with higher education presented higher Work Ability Index scores. Also, the age (B = −0.25, p < .001), work seniority (B = −0.19, p < .001) and education (masters' degree: B = 1.41, p = .012; ref. secondary) affect work ability. Conclusions The ageing process and seniority of nurses negatively affect their ability to work. A lack of programmes to maintain physical condition for nurses can result in a shortage of staff. Implications for nursing management Programmes can be developed to create or improve healthy working environments to increase productivity.

rates (Kruk et al., 2018). It was noted that each additional patient per nurse was associated with a 7% higher probability of death within 30 days of admission and a 7% higher chance of failure to rescue (Aiken et al., 2002).
In Poland, the shortage of nursing staff is caused, among others, due to low wages and emigration to countries with higher earnings (Marć et al., 2019). The fact is that the migration of the new generation of Polish nurses should now become a significant component of human resources policy in the Polish health service (Szpakowski et al., 2019).
It should be emphasized that one of the indicators accessing nursing care is the ratio of nursing personnel per 1,000 inhabitants.
In Poland, this indicator's value is 5.24, which is very low compared to Switzerland,Denmark or Germany,17.56,15.52 and 13.14, respectively (Abbasi et al., 2017). With the current dynamics of the population ageing, it is recommended that policies be introduced to raise the value to the average of OECD countries, which is 9.4 nurses per 1,000 inhabitants (Marć et al., 2019;Polish Chamber of Nurses & Midwives, 2017a, 2017bProniewicz, 2017).

Work ability is a fundamental social issue because it affects
workers' health and well-being (Mehrdad et al., 2016). Besides, the ability to work in a subjective assessment made by the employee results from the requirements of his work and the psychophysical abilities, which are personal characteristics.
Employers, more and more often, pay attention to the ability of their employees to work. With demographic changes increasing the proportion of older people in the population, employability is becoming crucial. Demographic projections in 27 European countries assume that over four decades -from 2010 to 2050the ratio of the post-working-age population to the working-age population (15-64) will double from 25.9% to 50.2% (European Commission, 2015).
Given the ageing population, which is becoming a global challenge for the health care sector, the increase in demand for nursing services will be closely linked to nurses' availability in the labour market. This means that nurses' physical and mental health should be at the highest possible level, translating into a high capacity to work. Nurses' ability to work is one of the topics discussed in the literature in the context of productivity when performing professional duties (Heyam et al., 2018).
It was decided to conduct a survey study among Polish nurses on the ability to work mainly due to the high workload (the result of the shortage of nurses in the labour market) and the ageing process.
It was investigated how the ability to work is affected by age, work seniority and nurses' level of education. The literature review shows that the ability to work may vary according to age group (Hatch et al., 2018) or education level (Golubic et al., 2009).
This study points out the decreasing ability to work among Polish nurses and indicates the need for multidimensional actions to prevent the total paralysis of nursing care, which can be faced by many European countries with a dynamically ageing population. Therefore, the study aimed to assess the ability to work in a professionally active group of nurses who are becoming ageing.
It has been hypothesized that the index of work ability decreases with nurses' age.

| Design and settings
The observational study was carried out from July 2018 to August

| Study participants
The criterion for inclusion in the study was the active employment of the respondents as a registered nurse. The study initially included 360 nurses, 11 of whom were excluded because they failed to complete the questionnaire. Finally, the study group consisted of nurses working in surgical and conservative wards. The size of the study group was 349 participants (347 female and 2 male).

| Data collection
The questionnaires were distributed to the nurses during their briefing about the research in the above-mentioned hospitals' conservative and surgical wards at the beginning of July 2018. Participation was voluntary and anonymous. After completing the questionnaire, each nurse placed it in an envelope and turned it in at the agreed location. At the end of August 2018, the questionnaires were collected and statistically analysed.

| Research tool
Data were collected using the WAI questionnaire, which is used to assess health care professionals' level of work ability. The questionnaire was adapted for use in Poland by Pokorski (1998). Through a self-assessment questionnaire, the WAI measures a worker's work ability and helps define necessary measures for maintaining and promoting work ability, as it helps to detect work-related health risks as early as possible. Appropriate action can then be taken, including occupational health care, to prevent declining capacity and early retirement.
The WAI results from the research work of Finnish Institution of Occupation Health (FIOH) employees (Tuomi et al., 1998). Their work aimed to develop a tool that could be used in occupational medicine to maintain work ability. The WAI expresses the subjective assessment of the worker's fitness for work. This tool was used by FIOH many times. In longitudinal studies, WAI allowed predicting changes in the ability to work in different professional groups. Also, the WAI provides a basis for future corrective and preventive action to improve or maintain employability.
The WAI questionnaire includes the following dimensions of the individuals: their current ability to work in comparison with their best years of life; their ability to work concerning their demand for work; the number of diagnosed diseases or limitations from which they suffer; their estimated impairments due to diseases/abilities or limitations; the number of sick leaves they have taken during the last year; the projections of their ability to work for 2 years (Juszczyk et al., 2019). Previous studies confirmed the validity and reliability of the WAI tool (Abbasi et al., 2016;Abdolalizadeh et al., 2012;Ebener & Hasselhorn, 2019;Peralta et al., 2012).
The WAI score ranges from 7 to 49 points and determines the employee's perception of work ability. The point value allows classifying the level of individual ability to work into the appropriate category and specify the required preventive and intervention measures' objectives. The score of 7-27 points corresponds to low work ability -restore work ability; a score 28-36 points corresponds to moderate work ability -improve work ability; a score of 37-43 points corresponds to good work ability -support work ability; finally, a score of 44-49 points corresponds to excellent work ability -maintain work ability (Adel et al., 2019).

| Ethical considerations
The research project was approved by the independent Bioethics Committee of the Wroclaw Medical University (no. KB-240/2018 and KB-426/2018). All participants gave written informed consent after a thorough explanation of the procedures involved. The study was carried out following the tenets of the Declaration of Helsinki.

| Statistical analysis
Sample size analysis was performed in Statistica 13 (TIBCO Software Inc., USA). Based on the results obtained in the hospital, the sample size was assessed in a pilot study. It was assessed how the ability to work in a professionally active group of nurses differed according to age. The pilot study was conducted among nurses aged under 30 and over 30. Means and standard deviations of WEI results in both groups were used to estimate the sample size. The estimated sample size for a two-sample unpaired-means test (unpaired t test).
Parameters: mean in the youngest group was 40.4 points (SD = 6.1); the mean in the oldest group was 38.1 points (SD = 8.9); the alpha level was set at 0.05 and the power of the test at 0.9. It also assumed no correlation of evaluated variables and adopted a 2-sided null hypothesis. The estimated sample size has been obtained equal to 316 people in the study based on the parameters. Also, the risk of losing patients in the follow-up assessment (20%) was assumed. The final sample size equals 348 participants. The final sample size in this research was 349 participants.
The data obtained during the survey were collected and systematized using Microsoft Office Excel spreadsheet tools. The statistical analysis was conducted using Statistica 13 (TIBCO Software Inc., USA) under licence from the Wroclaw Medical University, Poland. For measurable variables, arithmetic means, medians, standard deviations, and the range of variability (extreme values) were calculated. For qualitative variables, the frequencies (per cent) were calculated. All investigated quantitative variables were checked with the Shapiro-Wilk test to establish the type of distribution. Comparative analysis was performed using non-parametric tests -ANOVA Kruskal-Wallis rank with the post hoc test or Mann-Whitney U test. The correlation between the examined variables was performed using Spearman's rank correlation coefficient (r s ) .
An analysis of the impact of selected factors on work ability assessment was performed using linear regression (one-factor model of predictors included in the analysis). A non-standardized and standardized regression coefficient, standard error and level of statistical significance were determined. The next step was constructing the multi-factor model (progressive step method), taking into account the variables whose p-value in the univariate model was less than or equal to 0.50. For the overall study, a value of p < .05 was considered statistically significant.

| Basic characteristics
The mean age of the studied nurses was 46.9 ± 9.7 years (median 48), and the mean length of service was 23.5 ± 9.6 years (median 26). Over 99% (n = 347) of the participants were female. Regarding age group distribution, the highest percentages were recorded in the age groups of 41-50 and 51-60 years: 42.7% and 34.4%, respectively (Table 1). The group characteristics also included education level, work seniority, Work Ability Index (WAI), work position, work in more than one place and hospital ward. Detailed information on the characteristics of the studied group is presented in Table 1.

| Comparison of results of WAI depending on selected variables
Statistically significant differences (p < .001) in the WAI score were found between ages 21-30 and ages 41-50, 51-60 and over 60, as well as between ages 31-40 and ages 41-50, 51-60 and over 60 ( Table 2). The analysis of the WAI results showed a statistically significant correlation: the ability to work decreases with age (r s = −0.32, p < .001). In the youngest age group (21-30), the WAI median's value was 42 points, which corresponds to the category of good ability to work. In the age group of over 60, the median WAI dropped to 31.5 points, which corresponds to the category of moderate work ability ( Table 2).
The analysis of the relationship between the value of the WAI and the nurses' education level showed that statistically significant relationships exist between secondary education and bachelor's degree and between secondary and graduate education (p < .001; Table 2). This means that nurses with higher education scored higher WAI scores than nurses with secondary education.
Another statistically significant dependence showed that with increasing seniority, the ability to work decreases (r s = −0.26; p < .001). The analysis of the relationship between the WAI and the work position of nurses, work in more than one place, hospital ward did not show statistically significant differences (p > .05) ( Table 2). Table 3 contains the impact of selected parameters on work ability (WEI scores). The following variables were included in the analysis:  (Table 3). The observed demographic tendency towards an increase in the proportion of older people (i.e. those aged 60-65 and over) is, of course, not limited to Poland; it is a reality that many other countries face both Europe and worldwide. In Europe, Scandinavian countries -most notably Sweden -exhibit the highest ratios of older people (Bakalarczyk, 2012). In Poland, we can also observe a dynamically increasing percentage of older people. The continuing trend of population ageing will directly impact health, nursing and care services shortly.

| D ISCUSS I ON
Poland is suffering from an acute shortage of nursing staff. This  Polish study survey. This trend is also observed in other counties around the world (Halcomb et al., 2018;Spiva et al., 2011).
It should be emphasized that urgent actions are needed to reduce the shortage by making the profession more attractive to young people in Poland. Another important problem described by Szpakowski et al. (2019) should be considered due to the chronic shortage of health workers and the migration of nurses, which may be crucial in the near future, but this migration from Poland is not monitored in a reliable way.
Statistical analyses aside, it is self-evident that ageing diminishes the ability to work. This is a natural consequence of decreased psychophysical capacity, which may manifest in slower reaction times or increased cardiovascular and musculoskeletal  promoting healthy lifestyles and providing appropriate training (Abbasi et al., 2017;Silva et al., 2016).
Nurses working in inpatient facilities were analysed in light of the factors as mentioned above. The average seniority in the study group was 23.5 ± 9.6 years. It should be noted that the nursing pro-  Camerino et al. (2006) and Carel et al. (2013) have shown that the WAI score decreases with the age of the worker and that the factor with the most significant impact is the physical workload. The relevant literature emphasizes that people who score low on the WAI are more likely to leave the labour market sooner than those who score high. Significant independent factors causing premature total loss of the ability to work (pension transition) include high workloads, repetitive activities, uncomfortable postures, long working hours and shift work (Habibi et al., 2012;Makowiec-Dąbrowska et al., 2008).

Considering the ageing of the occupational group of nurses
and their shortage in the health care system, one should expect that they would prolong their employment after reaching retirement. That is not the case; however, Bugajska et al. (2011) and Heyam et al. (2018) stress that the ability to work of older people is adversely affected, among others, by shift work -especially at night, working overtime, exposure to chemical factors or heat, physical strain, working in multiple workplaces, the irregular pace of work and occupational stress. Eliminating these factors from jobs occupied by older people can have a positive impact on maintaining their ability to work. Milosevic et al. (2011) emphasize that nurses' satisfactory ability to work directly impacts their quality of life: the higher their ability to work, the better their physical and mental well-being and social relations.

| Study limitations
The study has several limitations. The study group is highly feminized, which reflects the fact that females dominate the nursing profession. Therefore, it was impossible to compare the ability to work between the two genders. Given the physiological and psychological differences between women and men, such a comparison might reveal differences in work ability. Future studies should ensure that a more representative sample of both sexes is selected.
Another limitation may have been the difficulty of completing the questionnaire: some questions may have been hard to understand for some respondents. A more advantageous solution would be to ensure the researcher's participation in the questionnaire completion by each respondent. The WAI seems to be a good source of information on potential limitations to the exercise of daily professional duties.

| IMPLI C ATI ON S FOR N UR S ING MANAG EMENT
Regarding nursing practice, based on the results of this study, programmes can be developed to create or improve healthy working environments that will help prevent occupational diseases and consequently increase productivity. Such measures can also serve to reduce workload and the ensuing fatigue and to increase employee motivation. This study showed that nurses with shorter work experience have a higher ability to work, and this may be due to the fact the workload was higher for those nurses working longer. The lack of development and implementation of a preventive programme for nurses would enable them to maintain good physical fitness levels, translating into an increased ability to work, resulting in a severe shortage of nursing staff in the near future.

| CON CLUS IONS
The nurses surveyed showed a moderate ability to work, which is related to the ageing of this occupational group. This is a significant challenge for the health care system, given that the largest group of nurses was between 41 and 60 years of age. Also, it was observed that nurses with shorter work experience have a higher ability to work.

ACK N OWLED G EM ENTS
There were no other contributors to the article than the authors as well as there was no writing assistance regarding our paper.
The certificated English language services were provided. All coauthors have agreed to the submission and publication of this manuscript.

CO N FLI C T O F I NTE R E S T
No conflict of interest has been declared by the authors.

AUTH O R S CO NTR I B UTI O N S
ŁR, IW and JR made substantial contributions to conception and design. ŁR was responsible for data collection. IW, PK and AK analysed data. ŁR and IW drafted the manuscript. JR, PK and AK revised manuscript critically for important intellectual content. All authors have participated sufficiently in the publication process to guarantee its content and take full public responsibility for the reported study, including its findings and conclusions. The final version is approved to be published by all authors.

E TH I C A L A PPROVA L
The research project was approved by the independent Bioethics